Reactions of Thiyl Radicals with Transition-Metal Complexes

employing DMPO (5,5-dimethyl-l-pyrrohe N-oxide) and TMPO ... metallic substrates, such as boranes, phosphines, and phosphites. ..... 0. 4.0. 8.0. 12. ...
2 downloads 0 Views 969KB Size
J . Am. Chem. SOC.1992, 114, 9510-9516

9510

Reactions of Thiyl Radicals with Transition-Metal Complexes Patrick Huston, James H. Espenson,* and Andreja Bakac* Contribution from the Ames Laboratory and the Department of Chemistry, Iowa State University, Ames, Iowa 5001 1. Received May 18, 1992

Abstract: Reactions of thiyl radicals (RS') were studied in aqueous solution by laser flash photolysis. Alkyl radicals, which were used to generate the thiyl radicals, were produced by the photohomolysis of the cobalt-carbon bond of RCo( [ 141aneN,)(H20)2+complexes (R = CH3,C2H5). These carbon-centered radicals abstract hydrogen atoms from thiols (C2H5SH, CysSH, or GSH), generating thiyl radicals. Thiyl radicals readily oxidize a variety of reduced transition-metal complexes. Kinetic and product studies were carried out, particularly for the ethylthiyl radical, to elucidate the mechanisms involved. Reactions of ethylthiyl radicals with inner-sphere reagents lead to metalsulfur bond formation, and reactions with outer-sphere reagents produce thiolate ions and oxidized metal complexes. Ethylthiyl radicals also react with RMLS2+complexes (M = Co or Cr; L = H 2 0 or tetraazamacrocycles). When R = CZHS,the thiyl radicals abstract a 8-hydrogen atom, generating ethylene as a product. However, when R = CH3, the reaction is an SH2substitution at carbon and C2H5SCH3is formed.

Introduction Thiyl radicals are intermediates in the oxidation of thiols to disulfides. These sulfur-centered radicals have been shown to be produced in biological systems in the reactions of oxidative enzymes with thiols.' The technique of pulse radiolysis has most often been used to generate thiyl radicals for chemical studies. The reactions responsible are those of H' or 'OH with thiols (eqs 1 and 2). With alcohols present, hydroxyalkyl radicals are H'

+ RSH

+ RSH

+

H2 + RS'

+ RS' 'OH + CH30H H2O + 'CH20H 'CH20H + RSH a CH30H + RS' *OH

-m

H20

+

(1)

(2)

RSSR

+ hv

(4)

+ hv

+

2RS'

(5)

R' + RSS'

(6)

+

RS'

(3)

generated (eq 3). They abstract a hydrogen atom from a thiol in the so-called repair reaction (eq 4)?.' This reaction is utilized in biological systems to protect against radiation damage and against naturally occurring radicals. Studies of the repair reaction involving a variety of carbon-centered radicals with thiols such as cysteine (CysSH), glutathione (GSH), cysteamine, and mercaptoethanol have been carried out previously by pulse radiolysis. The repair reaction typically has rate constants k4 on the order of 107-109 L mol-' The sulfur-hydrogen bond dissociation enthalpy in alkanethiols is 88 kcal mol-' and is independent of the length and configuration of the alkyl chain? Thus, alkylthiyl radicals are able to abstract hydrogen in the reverse of the repair reaction. This reverse reaction generally occurs much more slowly, with k4 typically in the range of 103-104 L mol-' s-l, depending on how strongly the C-H bond is activated.5 Thiyl radicals have also been generated by direct UV photolysis of disulfides. However, this may lead to cleavage of either the S-S or a C-S bond (eqs 5 and 6). For simple (methyl or ethyl) RSSR

of a C S bond is apparently not very important unless particularly stable alkyl radicals are products.6 Thiyl radicals have been identified by spin trapping studies employing DMPO (5,5-dimethyl-l-pyrrohe N-oxide) and TMPO (3,3,5,5-tetramethyl-l-pyrrolineN-oxide).' The lack of nuclear spin in the main sulfur isotope, 32S,precludes obtaining any structural information from ESR. Thiyl radicals are not easily detectable by optical spectroscopy, having molar absorptivities of only several hundred around 300 nm? For this reason, studies involving RS' have often been carried out in alkaline media (pH > 8), where the thiol is deprotonated and the highly colored disulfide radical anion is formed (eq 7).*-1°

disulfides, reaction 6 was found to be unimportant; the cleavage (1) (a) Foureman, G. L.; Eling, T. E. Arch. Biochem. Biophys. 1989,269, 55. (b) Wariishi, H.; Valli, K.; Renganathan, V.; Gold, M. H. J . Biol. Chem. 1989, 264, 14 185. (2) von Sonntag, C. In Sulfur-Centered Reactive Inrermediotes in Chemistry and Biolonv; - Chatgilialoglu, - C., Asmus, K.-D., Eds.;Plenum: New York, 1990; p 359. (3) Asmus, K.-D. Methods Enzymol. 1990, 186, 168. (41 Griller, D.; Simks. J. A. M.; Wavner. D. D. M. In Sulfur-Centered Reacfive Intermediates in Chemistry andEiologv; ChatgilialogL, C., Asmus, K.-D., Eds.; Plenum: New York, 1990; p 37. (5) (a) Schoneich, C.; Asmus, K.-D.; Dillinger, U.; von Bruchhausen, F. Biochem. Biophys. Res. Commun. 1989,161,113. (b) SchBneich, C.; Bonifacic, M.; Asmus, K.-D. Free Radicol Res. Commun. 1989, 6, 393.

+ RS- * RSSR'-

(7)

Thiyl radicals are known to oxidize compounds such as ferrocytochrome c, NADH, ascorbate, phenothiazines, and organometallic substrates, such as boranes, phosphines, and phosphites." The reduction potential for the RS'/RS-couple has been estimated for 8-mercaptoethanol to be 0.75 V vs NHE.I2 Recently, two studies in which thiyl radicals oxidized complexes of Cu(1) have been reported.13 Cysteinyl radicals oxidize the Cu1L2complex (L = cysteine) with a nearly diffusion-controlled rate constant ( k = 1.8 X lo9 L mol-' s-I). These studies were complicated by an equilibrium with CuzL3(eq 8), by acid-base equilibria, and by back reactions such as reaction 9. 2CuL2 F? CU2L3 + L RSSR'-

+ CuIIL,

-

RSSR

+ CuIL,

(8)

(9)

The presence of sulfur in the active site of many enzymes and its role in electron-transfer reactions make such studies important. Transition-metal-thiolate complexes may serve as models in un(6) (a) Morine, G. H.; Kuntz, R. R. Photochem. Photobiol. 1981,33, 1. (b) Byers, G. W.; Gruen, H.; Giles, H. G.; Schott, H. N.; Kampmeier, J. A. J. Am. Chem. SOC.1972, 94, 1016. (7) (a) Davies, M. J.; Forni, L. G.;Shuter, S.L. Chem. Biol. Inrerocr. 1987.61, 177. (b) Buettner, G.R. FEBS Leu. 1985,177,295. (c) Harman, L. S.;Mottley, C.; Mason, R. P. J . Biol. Chem. 1984, 259, 5606. (8) (a) Hoffman, M. Z.; Hayon, E. J . Am. Chem. Soc. 1972, 94, 7950. (b) Quintiliani, M.; Badiello, R.; Tamba, M.; Esfandi, A.; Gorin, G. Int. J . Radiot. Biol. 1977, 32, 195. (9) Purdie, J. W.; Gillis, H. A,; Klassen, N. V. Con. J . Chem. 1973, 51, 3 132. (10) Karmann, W.; Granzow, A.; Meissner, G.; Henglein, A. Inr. J . Rodiat. Phys. Chem. 1969, 1, 395. (11) (a) Forni, L. G.; Willson, R. L. Biochem. J . 1986, 240, 905. (b) Forni, L. G.; Willson, R. L. Biochem. J . 1986, 240, 897. (c) Forni, L.G.; Willson, R. L. J. Chem. Soc., Perkin Trans. Monig, J.; Mora-Arellano, V. 0.; II 1983, 961. (d) McPhee, D. J.; Campredon, M.; Lesage, M.; Griller, D. J . Am. Chem. SOC.1989, 111, 7563. (12) Surdhar, P. S.;Armstrong, D. A. J. Phys. Chem. 1987, 91, 6532. (13) (a) Leu, A.-D.;Armstrong, D. A. J . Phys. Chem. 1986,90, 1449. (b) Mezyk, S. P.; Armstrong, D. A. Can. J. Chem. 1989, 67, 736.

0002-7863192115 14-9510S03.00/0 0 1992 American Chemical Society

J . Am. C h e m . Soc., Vol. 114, No. 24, 1992 9511

Reactions of Thiyl Radicals w i t h Transition M e t a l s

derstanding the metal-thiol interactions in systems like non-heme iron-sulfur proteins.I4 We have recently reported a convenient method for generating and studying reactions of thiyl radi~a1s.l~ Studies of these radicals with a variety of aqua- and organotransition-metal complexes are described here. Experimental Section Materials. All water used in this study was in-house distilled, deionized water, passed through a Millipore-Q purification system. All chemicals were used as received unless noted below. Solutions were degassed by being purged with water-saturated argon (99.99% pure, Air Products Corp.). Ethanethiol (Johnson Matthey Electronics) was purified by being passed through a column of neutral activated alumina (Brockman activity of 1, 80-200 mesh, Fisher). The following complexes were prepared by literature methods: [RCo([ 14]aneN4)(H20)](C104)2 Ti(H20)63t,18 (R = CH,, C2H5, CD,, C2D5),I6 V(H2O)?','' ( H 2 0 ) 2 C r ([ 151aneN4)2+,19[Co(sep)] C13,20 ( R , R , S , S ) [Ni( [ 141 aneN4)](C104)2,21 (H20)5CrC2H52+,22 C2HSCr([ 15]aneN4)(H20)2+,19 and [Co(Me6[14]4,1 1[Co(C-meso-Me6[ 14]aneN4)(H20)2](C104)2, dieneN4)(H20)2](C104)2.23 Caution! Perchlorate salts of transitionmetal complexes, especially those of cyclam, are potentially explosive. Special care must be taken when working with these complexes. Stock solutions of N,N,N',N'-tetramethyl-1,4-phenylenediamine (TMPD) (Aldrich) were prepared by addition of the solid to degassed water and were protected from light. Solutions of Cr(H20)?+ in aqueous perchloric acid were prepared by reduction of solutions of Cr(C104)3over zinc amalgam. Solutions of Fe(H20)?+ were prepared in dilute aqueous perchloric acid, degassed, and placed over zinc amalgam. The concentrations of the solutions were determined spectrophotometrically by addition of an aliquot of the solution to exl,l0-phenanthroline (Fisher), ~ of forming F e ( ~ h e n ) ~(e510 ~ ' = 1.1 1 X lo4 L mol-' ~ m - ' ) . ~Solutions Ru(NH))~(C~O were ~ ) ~prepared by dissolution of the trivalent complex in dilute perchloric acid, removal of oxygen with an argon purge, and reduction over zinc amalgam. Concentrations were determined spectrophotometrically at 275 nm (e = 620 L mol-' ~ m - ' ) . Solutions ~~ of trans-Co( [ 14]aneN4)(H20)pwere prepared by mixing anaerobic solutions of C0(C104)2 and [14]aneN4 (Aldrich). This was stirred until complex formation was complete (ca. 10 min), acidified to 0.05 M HCIO,, and transferred without contact with air to zinc amalgam. The complex was stored in ice and used within 1.5 h. Concentrations were determined spectrophotometrically at 460 nm (c = 21.5 L mol-' cm-1).26 The complex was added to cells buffered at pH 7.0 immediately before the flash photolysis experiment to minimize isomerization to the cis isomer.27 Solutions of vitamin B12rwere prepared by zinc amalgam reduction of vitamin B12ain neutral solution and used immediately.28 Thiolatochromium(II1) complexes of the formula (H20)&rSRZ+were prepared and purified using several different methods. An acidic aqueous solution containing Cr(H20)?+ and excess diethyl disulfide was photolyzed at 254 nm using a Rayonet photochemical reactor with mediumpressure mercury lamps. The reaction mixture was placed on a cooled, degassed column of Sephadex C25 cation exchange resin. However, despite repeated washings, the purified sample, eluted with 0.5 M NaC104 in 0.01 M HC104, still contained disulfide. The second method was similar to the laser experiments. Similar concentrations of CH,Co([ 14]ar1eN~)~+, C2H5SH,and Cr(H20)s2t were used, but the scale was -15 times larger. Upon cation exchange, the pure yellow thiolato-

-

(14) Malkin, R.; Rabinowitz, J. C. Ann. Rev. Biochem. 1967, 36, 113. (15) Huston, P.; Espenson. J. H.; Bakac, A. Inorg. Chem. 1992,31,720. (16) Bakac, A.; Espenson, J. H. Inorg. Chem. 1987,26,4353. [14]aneN4 = 1,4,8,1I-tetraazacyclotetradecane(cyclam). (17) Espenson,J. H.; Bakac, A.; Kim, J.-H. Inorg. Chem. 1991,30,4830. (18) Bakac, A.; Orhanovic, M. Inorg. Chim. Acta 1977, 21, 173. (19) Samuels, G. J.; Espenson, J. H. Inorg. Chem. 1979, 18, 2587. [ 15]aneN, = 1,4,8,12-tetraazacyclopentadecane. (20) Creaser, I. I.; Geue, R. J.; Harrowfield, J. M.; Herlt, A. J.; Sargeson, A. M.; Snow, M. R.; Springborg, J. J. Am. Chem. SOC.1982, 104, 6016. C~(sep)~'= (S)-( 1,3,6,8,10,13,16,19-octaazabicyclo[6.6.6]eicosane)cobalt(II). (21) Bosnich, B.; Tobe, M. L.; Webb, G. A. Inorg. Chem. 1965,4, 1109. (22) Hyde, M. R.; Espenson, J. H. J . Am. Chem. SOC.1976, 98, 4463. (23) Rillema, D. P.; Endicott, J. F.; Papaconstantinou, E. Inorg. Chem. 1971, 10, 1739. C-meso-Me,[ 14]aneN4 = C-meso-5,7,7,12,14,14-hexamethyl-1,4,8,1l-tetraazacyclotetradecaneand Me6[14]4,11-dieneN, = 5,7,7,12,14,14-hexamethyl-1,4,8,11-tetraazacyclotetradeca-4,1 I-diene. (24) Ford-Smith, M. H.; Sutin, N. J . Am. Chem. SOC.1961, 83, 1830. (25) Annor, J. N.; Scheidegger,H. A.; Taube, H. J. Am. Chem. Soc. 1968, 90, 5928. (26) (a) Bakac, A.; Espenson, J. H. Inorg. Chem. 1989, 28, 3901. (b) Heckman, R. A.; Espenson, J. H. Inorg. Chem. 1979, 18, 38. (27) (a) Poon, C. K.;Tobe, M. L. J. Chem. SOC.( A ) 1968, 1549. (b) Tsintavis, C.; Li, H.-L.; Chambers, J. Q.Inorg. Chim. Acta 1990, 171, 1. (28) Balasubramanian, P. N.; Pillai, G. C.; Carlson, R. R.; Linn, D. E., Jr.; Gould, E. S. Inorg. Chem. 1988, 27, 780.

chromium(II1) complex was eluted with 0.5 M NaC104 in 0.01 M HCI04. This method allowed small amounts of CrSR2+to be obtained pure. The last method employed was the reaction29of Cr(H20)?+ with (CH3)2CHSCo(dmgH)2, yielding (H20)5CrSCH(CH3)22+. The 2propanethiolatocobaloximewas prepared by a variation of the procedure for the MeS and PhS derivatives.'O This method allowed preparations of CrSR2+ to be carried out on a much larger scale. AMIYM. Chromium analyses were carried out on the thiolatochromium(II1) complexes by oxidation to chromate (6372 = 4830 L mol-' cm-I) in basic peroxide.)' Cobalt(I1) was determined as C O ( N C S ) ~ ~ (e623 = 1842 L mol-' ~ m - 9 .Ethanethiol ~~ and ethyl methyl sulfide were detected by the use of a Hewlett-Packard 5790A gas chromatograph with a Porapak Q column at 170 OC; methane, ethane, and ethylene were detected with a VZ-10column at 50 OC. The ratio of sulfur to chromium was determined by the use of inductively coupled plasma mass spectrometry (ICP/MS). First, the complex (H20)5CrSC2H52+was prepared by photolyzing a solution containing CH,Co( [ 14]a11eN,)~~,C2H5SH,and Cr2+with visible light, as described above. The product mixture was placed on Sephadex C-25 cation exchange resin and eluted with 0.10 M LiBr/2 mM HBr. Sodium ions must be avoided because of the space charge effect; chlorine must be avoided because its mass-to-charge ratio is similiar to that of sulfur. The in~trument'~ was a Sciex Elan Model 250 (Perkin-Elmer). The sample was introduced to the plasma by means of a Cetac U-5000 ultrasonic nebulizer. Sulfur was monitored at a mass-to-charge ratio of 34 and chromium at both 52 and 53. Laser Experiments. Kinetic studies were carried out in 0.10 M HC104 except where noted. Reactions of the sulfur-centered radicals were measured by the use of a visible dye laser flash photolysis system described previ~usly.'~ The excitation dye used was LD 490 (Exciton). Increases in absorbance due to formation of ABTS'- were monitored at ) ~ ~those due to TMPD'+ at 650 nm (e = 1.20 X lo4 L mol-' ~ m - ' and ~ ~ or the other kinetic probe 565 nm (e = 1.25 X lo4 L mol-' ~ m - ' ) . One was used in all the experiments carried out by use of the dye laser. The Nd:YAG laser system used here was an LKS.50 laser photolysis spectrometer from Applied Photophysics Limited. The laser itself was an SL800 system from Spectron Laser Systems. The fundamental wavelength output from the Nd:YAG laser (Q-switched) was frequency quadrupled by propagation through nonlinear harmonic generating crystals, yielding 266-nm light. The laser beam was set up perpendicular to the monitoring beam, a pulsed xenon lamp. The monitoring beam passed through the cell and through a grating monochromator to a five-stage photomultiplier tube. The signal was recorded on a PM3323 Philips digital oscilloscope interfaced to an Archimedes 420/ 1 computer. Using this system, diethyl disulfide was photolyzed at 266 nm in the presence of Cr(H20):'. The formation of (H20)JrSR2+ was monitored directly at 280 nm without the use of a kinetic probe. The rate law for the loss of radical in the presence of substrate, which was present in pseudo-first-order excess, is given by eq 10. However, -d [ RS'] -dt

- 2kI7[RS'l2 + k,[substrate] [RS']

(10)

-

as described p r e v i o ~ s l y data , ~ ~ were ~ ~ ~ fitted ~ to a first-order equation, as the production of only 1 X lod M radical in a laser flash made the second-order component small. The validity of this approximation was confirmed using the kinetic simulation program KINSIM.38 Three of the data sets were also subjected to a rigorous mixed first- and secondorder kinetic analysis. The second-order rate constants k, obtained by the two methods agreed within 10%. In two cases, the second stage, attributed to the reduction of ABTS'by reduced metal complex (see below), had rates comparable to those of (29) Lane, R. H.; Sedor, F. A.; Gilroy, M. J.; Bennett, L. E. Inorg. Chem. 1977, 16, 102. (30) Schrauzer, G. N.; Windgassen, R. J. J . Am. Chem. SOC.1967, 89, 3607. dmgH = 2,3-butanedione dioxime (dimethylglyoxime). (31) Haupt, G. W. J. Res. Natl. Bur. Stand. 1952, 48, 414. (32) Kitson, R. E. Anal. Chem. 1950, 22, 664. (33) Smith, F. G.; Houk, R. S. J. Am. Soc. Mass Spectrom. 1990, 1, 284. (34) (a) Hoselton, M. A.; Lin, C.-T.; Schwarz, H. A.; Sutin, N. J . Am. Chem. SOC.1978,100, 2383. (b) Melton, J. D.; Espenson, J. H.; Bakac, A. Inorg. Chem. 1986, 25, 4104. (c) Connolly, P.; Espenson, J. H.; Bakac, A. Inorg. Chem. 1986, 25, 2169. (35) Hiinig, S.; Balli, H.; Conrad, H.; Schott, A. Liebigs Ann. Chem. 1964, 676, 36. (36) Fujita, S.;Steenken, S . J . Am. Chem. SOC.1981, 103, 2540. (37) (a) Bakac, A.; Espenson, J. H. Inorg. Chem. 1989, 28, 4319. (b) Kelley, D. G.; Espenson, J. H.; Bakac, A. Inorg. Chem. 1990, 29, 4996. (38) Barshop, B. A.; Wrenn, R. F.; Frieden, C. Anal. Biochem. 1983,130, 134. We are grateful to Professor Frieden for a copy of this program.

9512 J. Am. Chem. SOC.,Vol. 114, No. 24, 1992

6

-

5

-

Huston et al. Table I. Rate Constants (-25 "C) for Reactions of Thiyl Radicals with A B T P (2,2'-Azinobis(3-ethylbcnzothiazoline-6-sulfonate) iona RS' klS/i07L mol-I s-I PH

5.6 i 0.2 1.o 50 f 4 7.O CYSS' 78 f 2 4.0 1OOb 3.75 8.8 f 0.2 7.0 5@ 7.1 "The pH was adjusted with phosphate buffer (pH 7.0) or by addition of HC10,. Errors given are standard deviations in the data set as calculated by a nonlinear least-squares fitting program. Reference 39. CzH,S'

4 -

B

44

0.002 0

40

80

120

160

Time / lo6s 0

1

3

2

4

5

10 6

7

[ABTS27/lo-' M Figure 1. Plot of kob against the concentration of ABTS2- for the reaction of CysS' with ABTS2- in 0.05 M phosphate buffer (pH 7.0)at -25 OC. Inset shows the increase in absorbance at 650 nm with 5.04 X lo4 M ABTS2-. the first stage. Thus, data for the reactions of Fe(CN),& and Cr([15]a r ~ e N ~ were ) ~ + fitted to a biexponential equation which yielded values of k, and k l l . ABTS-

-

+ L6S"

ABTS2'

+ L6Mn+I

+ + + + R'*

C0([14]aneN~)~+ (12)

+ RS' R" R', or {R'(-H) + R'H) R" RS* ABTSZ- RS- + ABTSR"

RSH

R'H

(13) (14) (15)

water-soluble thiol present, such as ethanethiol, cysteine, or glutathione, the repair reaction was employed to generate thiyl radicals (eq 13). Because thiyl radicals are not highly colored,8 two kinetic probes were used. Thiyl radicals were allowed to react with 2,2'-azinobis(3-ethylbenzothiazoline-6-sulfonate) ion (AB=,-) in a known reaction (eq 15)39to yield the highly colored radical anion ABTS'-. Another probe used was TMPD, which is easily oxidized to the highly colored radical cation TMPD'+. Thiyl radicals were observed to oxidize TMPD in neutral solution but not in acidic solution, as the protonated amine is much less easily oxidized. Kinetics. The repair reaction was studied for methyl and ethyl radicals. With [AB=,-] >> [RSH] (lo-* and lo" M, typically), reaction 13 is rate limiting. Under these conditions, the observed rate constant is given by eq 16. The first term in eq 16 is due

kob = 2k14[R'*] + kl3[RSH]

(16) to carbon-centered radical self-reactionsNand is small (generally 6% of kob). The second term, k13[RSH],is due to the repair reaction. Thus, the pseudo-first-order rate constant is directly proportional to [RSH] and is independent of [ABTS2-1. A plot of kob vs [RSH] is linear, with a slope corresponding to the second-order rate constant k I 3and a small intercept due to radical self-reactions. The repair reaction was studied for 'CH3 with ethanethiol, cysteine, and glutathione and for T 2 H 5with ethanethi01.l~ Methyl radicals abstract hydrogen atoms from ethanethiol with second-order rate constants of (4.0 f 0.2) X lo7 (39) (a) Lal, M.; Mahal, H. S.Can.J . Chem. 1990,68, 1376. (b) Wolfenden, B. S.; Willson, R. L. J . Chem. SOC.,Perkin Trans. I I 1982, 805. (40) Stevens, G. C.; Clarke, R. M.; Hart, E. J. J . Phys. Chem. 1972, 76,

3863.

fv)

*2

6.0

4.0

2.0

(11)

ReSultS Generation of Radicals. One method for producing carboncentered radicals (- 1 X 10" M) is the visible (490 nm) laser flash photolysis of aqueous solutions containing R'Co( [ 141aneN4)(H20),+(typically 1 X lo4 M), as in eq 12.26a With a R'C0([14]aneN~)~+ + hv

8.0

0

4.0

8.0

12

16

20

[ML62+] / lo" M Figure 2. Observed pseudo-first-order rate constants at -25 OC and pH 1.0 (HClOJ, corrected for radical self-reactions and the reaction of C2HSS' with ABTS2-, varying linearly with [ C o ( ~ e p ) ~ +and ] [CrWz0)62+1.

and (4.7 f 0.2) X lo7 L mol-] s-l at pH 1.0 and 7.0, respectively. Ethyl radicals react with ethanethiol at pH 7.0 with k I 3= (2.8 f 0.1) X lo7 L mol-] s-l. Cysteine and glutathione have virtually identical reactivities toward 'CH, at pH 7.0 (kl3 = (7.4 f 0.2) X lo7 L mol-' s-' for CysSH and (7.1 f 0.2) X lo7 L mol-' s-I for GSH). When [ABTS*-] at pH 7.0. The second-order rate constant for this reaction as well as for the oxidation by ethylthiyl radicals is listed in Table I, along with the two values for cysteinyl radical from previous s t u d i e ~ . ' ~ JTMPD ~ was also oxidized by C2HSS*,forming TMPD'+ at pH 7.0, with a second-order rate constant of (2.6 f 0.1) x io9 L mol-l s-l. Knowledge of the rate constants for reactions of thiyl radicals with intensely colored kinetic probes allows the study of their reactions with weakly colored transition-metal complexes to be carried out by the competition method. Typical conditions used were as follows: 1 X lo4 M CH3Co([14]aneN4),+,which generated -1 X 10" M 'CH3 in the laser flash (eq 12); 0.10 M C2HsSH, to efficiently react with methyl radicals, generating M TMPD C2H5S' (eq 13); -6 X lo4 M ABTS2-or -2 X as a kinetic probe; and an appropriate amount of metal complex (eq 19) to make a measurable contribution to the observed

J . Am. Chem. SOC.,Vol. 114, No. 24, 1992 9513

Reactions of Thiyl Radicals with Transition Metals Table 11. Rate Constants (-25 "C) for Reactions of C2H5S'with Transition-Metal Complexes Obtained by Use of ABTSZ-as a

Kinetic Probe" kI9/Lmol-' s-I PH Cr(H20)p (4.9 t 0.2)x 108 1.0 V(H20)P (6.5f 0.3)X lo8 1.0 Fe(H20)p (1.2f 0.1) x 106 1.0 Cr( [ 15]aneN4)2t (2.2f 0.2) x 108 1.0 Co([14]aneN4)2+ -4.5 x 108 7.ob CO([Me6[14]aneN4)2+ (3.3 f 0.3)X lo8 7.0b C0(Me~[14]dieneN~)~+ (3.1 f 0.3)X lo8 7.0b vitamin B12r (1.0 0.1) x 109 7.p Ni([ 14]a11eN,)~+